Adaptive Multi-Rate (Adaptive Multi-Rate, AMR for short) is a speech codec scheme used in communication and is specifically divided into adaptive multi-rate narrowband codec (Adaptive Multi-Rate Narrowband, AMR-NB for short) and adaptive multi-rate wideband codec (Adaptive Multi-Rate Wideband, AMR-WB for short).
As for AMR-NB, a voice signal with a frequency bandwidth ranging from 300 Hz to 3400 Hz is generated by using an 8 KHz voice sampling frequency; and for AMR-WB, a voice signal with a frequency bandwidth ranging from 50 Hz to 7 kHz is generated by using a 16 kHz sampling frequency. For a low-frequency part from 50 Hz to 300 Hz, voice naturalness and a sense of presence are improved; for a high-frequency part from 3.4 kHz to 7 kHz, intelligibility of sounds and clearness of fricatives are increased. Therefore, compared with AMR-NB, AMR-WB can provide high-definition voices for a user.
To provide better experience in a voice service, an increasing quantity of terminals support the AMR-WB technology, and mobile telecommunications operators are gradually deploying high-definition voices that are based on the AMR-WB technology. Therefore, there exists, on an existing network, a phenomenon that not all devices are capable of providing an AMR-WB voice service.
In a process of implementing a voice service, it is required that all logical functional entities (for example, a core network, an access network, and a terminal) in an end-to-end link are capable of providing the AMR-WB voice service, so that an AMR-WB radio bearer can be set up to provide a high quality voice service. Therefore, quality of the voice service depends on a device that has a lowest voice service codec capability in a system. Between two parties (calling/called terminals) in communication, when a terminal at only one end provides the AMR-WB voice service while other devices provide only an AMR-NB voice service, a service capability is restricted, only a radio bearer for the AMR-NB voice service can be set up between the terminals, and both the two parties in communication use AMR-NB to implement a low quality voice service. Therefore, a user of the terminal that has an AMR-WB capability cannot experience a high-definition voice. As a result, user experience of the terminal is inconsistent with the capability of the terminal, and user experience is poor.